I would like to welcome everyone to the Akamai second quarter 2006 earnings call. (Operator Instructions) Ms. Smith, you may begin your conference.

Thank you. Good afternoon and thank you for joining Akamai's investor conference call to discuss our second quarter 2006 financial results. Speaking today will be Paul Sagan, Akamai's President and Chief Executive Officer; and JD Sherman, Akamai's Chief Financial Officer. Today's presentation contains estimates and other statements that are forward-looking under the Private Securities Litigation Reform Act of 1995. These statements are based on current expectations and assumptions that are subject to risks and uncertainties, and involve a number of factors that could cause actual results to differ materially. Additional information concerning these factors is contained in Akamai's filings with the SEC including our Annual Report on Form 10-K. While we may elect to update forward-looking statements at some point in the future, we specifically disclaim any obligation to do so, even if our estimates change. Therefore, you should not rely on these forward-looking statements as representing our estimates as of any date subsequent to today. During this call, we will be referring to some non-GAAP financial measures that we believe are helpful to a better understanding of our financial results and operations. These non-GAAP measures are not prepared in accordance with Generally Accepted Accounting Principles. You can find definitions of these non-GAAP terms and reconciliations of these non-GAAP terms to the most directly comparable GAAP financial measures under the News and Publications portion of the Investor Relations section of our website. Now let me turn the call over to Paul.

Thank you, Sandy. Thank you all for joining us today. We had an outstanding second quarter at Akamai. We experienced unprecedented demand for many segments of the business, allowing us to significantly overachieve our own expectations. Our results for the second quarter include revenue of $100.6 million, a $10 million increase over the first quarter. That's 11% sequential growth and a 56% increase over the second quarter of last year. We are especially pleased to have achieved our first $100 million quarter ahead of plan. Normalized net income was more than $35 million or $0.20 per diluted share. That's more than double our normalized net income of a year ago and 22% higher than our last quarter. While we carried some notable events in the quarter, our tremendous revenue growth was driven by strength across many customer segments and geographies. Enterprises are finding more and more ways to profit from increased online consumer activity. That means they are turning to us more and more often to ensure a high quality online experience for their target audiences across the globe. This is the powerful trend we talked about last quarter, and we've seen it continue accelerating through the first half of this year. I will have some more comments about the trends in our business later. But now, let me turn it over to JD to review our financial results in detail.

Thanks, Paul. As Paul said, we had an outstanding second quarter, delivering double-digit sequential revenue growth for the second quarter in a row and 22% normalized net income growth quarter-over-quarter. Revenue came in at $100.6 million, well above the high end of our expectation range. In addition to significant event traffic that we had expected, we also delivered some events that we hadn't anticipated at the time of our last earnings call, including some large software download events. But more importantly, we're seeing an increase in usage across our customer base, which is driving increased bursting which in turn drives larger commitments from our customers as they renew their contracts. We've seen this trend for a couple of quarters in a row now, and we believe we are benefiting from the continued proliferation of broadband and increased consumer use of the Internet. As we mentioned on our last earnings call, we are seeing an increase in larger longer-term deals, particularly from our media and entertainment customers and that continued this past quarter. For the quarter, our average revenue per customer, or ARPU, grew to $16,400 per month. That's up 7% on top of the 5% sequential growth we saw in the first quarter, demonstrating that growth from our existing customers has been driving a significant benefit to our top line. We also continue to add new customers to the platform, adding 79 net new customers in the second quarter, bringing our total customer count to 2,060. Sales were strong in all geographies. International sales represented 22% of total revenue, up from 21% in the same period last year but down slightly from last quarter. Resellers accounted for 22% of total revenue. Once again, no customer accounted for 10% or more of our revenue. Our GAAP gross profit margin was 79%…

Thanks, JD. As you just heard, we had a terrific quarter and our new guidance for 2006 makes this year the most promising in Akamai's history. We established what we thought were ambitious goals at the beginning of the year. Now, just halfway through 2006, it is evidenced that growth in our business has accelerated faster than we had anticipated and that's a good thing. This gives us the confidence to raise our full year outlook to $415 million to $420 million as JD just detailed. Our accelerated growth has been driven by higher than expected demand for our services across the business. We believe that broadband adoption is an important enabler of this trend. With broadband connections now in over half the online homes in the US, Forrester Research estimates that broadband users spend up to 18 hours a week online. That's almost three hours more per week than the average dial-up user. Not only are they online more, they do more online because of their fast connections. Broadband users also outspend their dial-up counterparts online by almost 40%, according to Forrester. That makes broadband users a more compelling target for profitable Internet businesses. We're witnessing the success of a variety of business models online. Transaction models online are working. This is being demonstrated by the continued double-digit growth of online retail sales year-over-year. Pay-per-view and subscription models are working online, led by the success of online music services like iTunes, which we support. And now TV shows and movie sales are starting online, which offers us new growth opportunities. Advertising models are working online with Forrester projecting online ad spending to grow from $13 billion in 2005 to $24 billion in five years. Many of the sites that rely on advertising revenue or sell and serve those ads…

(Operator Instructions). Ranjini Chandirakanthan, ThinkEquity.

Hi, good afternoon and congratulations on the great quarter. First question I have for you is on the new customer count, obviously growing again. I'm curious how much more is left of the greenfield opportunity here in the US? Have you seen any pressure or are you seeing some of the new customer count come from new businesses?

You know I don't think we see a limit on the market opportunity domestically or internationally yet. More and more businesses are going to the web and finding profitable models and that continues to expand. We believe the opportunity for us is to sell to new customers. In addition, some of the things that we've added, such as our application performance services, within the last year has opened some new markets like manufacturers, who traditionally you wouldn't think of as in the B2C web business but are very much in the B2B space, and care about the performance of their online portal. So we see lots of new opportunities. I think just as important was the very strong ARPU growth though, 7% increase quarter-over-quarter, because that indicates that our existing customers are doing more and more on the web and using more and more of our services. We try to balance both if you will, mining the existing customers particularly now that we have such a large set of great customers as well as finding the right new customers to bring onto the platform. You said you had a second question?

Yes. Continuing on the new customer count, should we expect to see more of those revenues coming next quarter? I guess it's a question of linearity.

We don't give you quarter-over-quarter customer count guidance. It always fluctuates a little bit. We don't have a target number; we would like to manage towards getting good customers, not just any number of customers. But the one thing in terms of how that layers on is we recognize revenue as customers actually go on the network and start pushing traffic or using features. So a customer, who is added early in one quarter, probably contributes full revenue for each of the three months in the following quarter. Someone, who joined late in the quarter probably only contributes a piece to the next quarter. Generally, customers come on pretty steadily through the quarter, so we get a pretty even distribution in new contribution quarter-over-quarter.

Thanks so much. Great quarter again.

Our next question comes from Katherine Egbert - Jefferies.

Hi, how are you? A couple questions. First, it seems like you had some bigger contracts this quarter. Can you tell us what percentage of revenue is bursting? Can you talk particularly in the media and entertainment divisions, I mean what kind of features, be it TV shows or movies, are you enabling there? Can you just tell us where you are at? Then give us some sense of how these customers grow over time, specifically in media and entertainment?

Let me try to pick up on all of those things. We certainly strive for bigger relationships with our customers. You see that in the ARPU growth and often that gets reflected then in the larger renewals. The mix of committed revenue versus bursting, which we have said is traditionally through several evolutions of the business. It's been kind of a 70/30 split, about 30% bursting. That held pretty much true yet again, so that was nice to see. In the M&E space, we're seeing it in really a wide variety of customers in geographies around the world. So we're seeing it in live video in the case of things like sporting events. We're seeing it in on demand in things like music in particular. We're starting to see some movies and TV shows sold. Of course, iTunes does that with some TV shows so that work that we do is quite visible. As that business grows, it's obviously very data intensive and so we get a disproportionate growth in potential business or traffic on the network. Because if you will, an average visit is a lot bigger if you are downloading a TV show than if you're visiting one random web page. So, we like to see growth in that area. We're seeing it both in live, and we're seeing it in on-demand. But frankly, we're seeing it not just in media companies but we're seeing it as other businesses richen up their websites. We had a very large event with a computer manufacturer in the past quarter, who did a live event online that drove a very large audience as part of a promotion of their product using some big-named talent. So we are seeing it in traditional media if you will -- if you can use the word traditional on the Internet -- and we're seeing it as other businesses effectively compete for the attention of eyeball. So we're seeing it across a wide variety of our customers, even outside of the media and entertainment category. Then they grow randomly from one site to the other. Someone has a bigger quarter than another. But generally, we're just seeing them grow as broadband usage has increased. There are a couple of trends there. One is just the sheer number of broadband users, and we passed that 50% threshold in the US. But we're also seeing it as broadband gets bigger over time, as people start maybe with a small broadband connection and they move up to a cable modem and eventually maybe to fiber. Their ability to again make an individual visit bigger by watching a high-resolution video or downloading an average file adds value to the visit and therefore to the services that we can provide our customers.

Thanks. That is helpful. Can you tell us how much the World Cup contributed?

No, we don't break out individual customers. That was a nice event. We were able to work with a number of customers because the rights there are secured sort of like the Olympics, by geography, so it's not just one customer. We work with one client in the US. We work with a different one in Brazil, for example, so it contributed to a number of customers. It was certainly an event we knew about going into the quarter, so we had some ability to estimate it. We didn't know the full extent or everything that all of our customers were going to do or how big it would turn out. Nice event, but it wasn't the single thing that really drove the over-performance in the quarter. It was usage and business across most of our verticals with a very wide selection of our customers, which is really terrific.

Then last question. Your new multi-year contract with Microsoft, does it cover the release of Vista when it goes into general availability?

I won't make any specific comments on the details of the contract except to say we're thrilled with the relationship. It goes back the life of our Company, practically. They were an investor. They've been one of our oldest customers. By the time this contract comes up, it's almost hard for me to comprehend, but they will have been a customer heading towards a decade because we will have been in business that long. So I was very pleased that they continue to work with us and we work hard for them.

Thanks, Paul. Congratulations.

Our next question comes from Jeff Van Rhee - Craig-Hallum.

Thanks and congrats, let me add to the list. A couple of questions. First of all, that's fantastic ARPU growth. To the degree that you can give us any color at all, is the vast majority of that being just flat-out driven by greater usage or are you seeing any material increases in terms of the functionality? The bells and whistles and different features that you offer that people are buying? Your contract length, you had commented you had started to see some contract lengthening last quarter. Curious if you have continued to see that trend or how that has changed, if at all.

I think the ARPU growth was a combination of both factors, which is what we like to see; just more volume from customers as well as up-selling new services and features, including application acceleration technologies. I'll let JD speak to customer length.

Yes, I think we're seeing contracts length in general get a little bit longer. We're averaging probably about 18 months on our new contract lengths; that had been closer to 14 or 15 months in the past. So the lengths are definitely getting longer and the sizes are getting bigger with some of our big customers, especially in the media and entertainment space.

Paul, on the ARPU, is 75/25 a reasonable way to think about volume of usage versus new features and functionality in terms of what's driving ARPU?

I don't think we've really been able to look at it that closely on the last quarter. You know it's been generally half-and-half. Maybe it's a little more on the volume side because of broadband; that would be my hunch. But we haven't looked at it that carefully obviously as we work through product management and release new offers. We will study that carefully.

In terms of the CapEx, it looks like more R&D being capitalized both as a percent of overall R&D and in terms of absolute dollars. Is that a trend we should expect to continue?

We have a little bit more capitalized R&D this quarter. A higher portion of our R&D effort was capitalizable. It's about 15% of total capital expenditures, and I expect it to be in that range.

Congrats.

Our next question comes from Michael Turits - Prudential.

A couple questions here. First, did you mention the amount of churn that you guys saw this quarter?

I'm sorry; what was the question?

Churn.

The amount of churn that you saw, customer churn?

We did not mention it. It was about 4%, about the same level that we saw in the last quarter.

Just to remind you that's really the last of the Speedera integration quarter, so there was still some of the very small customers coming off the network. So we still expect to keep it in the low single-digits for the rest of the year.

So some potential for that to go down a little bit more?

A little bit. But it's very low. And again, the customers who are coming off are much, much smaller.

Any comment in the media space, especially on growth in bandwidth, on trends there? We saw a big acceleration in the last year or so, I think possibly driven by the increase in video. Are we still getting from the bulk of your customers the same growth rates in bandwidth usage or is that beginning to trail off?

No, I think we see -- and again, I'm generalizing -- but very strong growth across many segments, not just M&E, but M&E is probably the leader there because there's so much audio and video going on. But we see it in the software distribution category. We see it among commerce customers, who are richening up their sites and adding video as well. So it seems to be a strong general trend.

So it doesn't seem as if the bandwidth growth rates have peaked yet. Is that what it seems?

You mean use of the network?

The growth rates, yes. The growth rates and the use of the network on a bandwidth basis.

I don't think so. Certainly, all of our modeling and stuff suggests that the curve should continue very strongly because there are more people using the Internet. A very small fraction of the world actually is connected. Not everyone has broadband by any means. In many cases, unfortunately actually in the US, we trail broadband compared to many parts of Asia. So there's a lot of upgrade that I think will come over the next five years in terms of just what it even means to have a broadband connection. So I think that those trends are very positive.

Last question for JD. Where do we get to a step function here on CapEx? You're investing now. When can we begin to see that flatten out as a percentage of revenues or more importantly decline as a percentage of revenues?

I think a lot of that will depend on where we see the opportunity. Because we always want to make sure that we have the capacity in place so that we can deliver for our customers in terms of scalability and just plain handling the capacity. If we keep seeing growth like this, we will keep investing for it. But I don't believe we're going to see a step function upward in our level of capital expenditures. I think what we're seeing now is an opportunity to really capture some more demand based on the marketplace, and we've made an incremental investment to do that.

Our next question comes from Mark Kelleher - Canaccord Adams.

I just wanted to focus a little bit on the software, the large software downloads you indicated. Should we look at that as maybe an unexpected one-time event? Is it connected with the renewal of the Microsoft contract? Associated with that, were there any 10% customers in the quarter?

There were no 10% customers, as JD had mentioned. We haven't had one in a long time. That's a good thing. We like a wide diversification of our customers by vertical as well as just individually. The software trend is really across the board. It's among people who are in the OS business, but it's also people who are in the software security and the antivirus space. So we continue to see that trend among a number of our customers, big and small.

So it wasn't specifically connected to the extension of the Microsoft contract?

We were renegotiating the contract. We are very happy with that result. We've been happy with Microsoft as an individual customer really for seven or eight years. We've got a great relationship there. We just continue to see a lot of growth in the software delivery space across many customers, not just any one in particular.

One more quick question. You touched on application acceleration. Could you just give us an update on how that is going? How that is rolling out?

Sure. We don't break out a lot of specifics and we did a first six-month update a little while ago. But we continue to see adoption. We continue to see, we think, a very good sales model and pricing model there with B2B customers as well as B2C customers. It's an area that we will continue to invest effort into developing a richer product portfolio and developing new business, and it's an area we're very committed to. Because what we see is that people have spent a lot of time and money trying to improve performance of applications hosted in data centers with appliance or hardware solutions, and that provides some benefit but does not solve a lot of the problems, particularly when you start to go across the public Internet and you can't control the Internet performance or all of the endpoints. It's one thing to worry about performance between two linked data centers. It's another to deal with performance to say several thousand partners or suppliers. That's where our overlay solution as a total solution married to content delivery, we think is very unique and a differentiator and where we continue to get traction in the marketplace and will continue to put some of our effort.

Great, thanks. Congrats on the quarter.

Our next question comes from Tim Klasell - Thomas Weisel Partners.

Hi guys, congratulations. The first question is on your CapEx. Can you give us an idea of what sort of projects you have laid out? Are there any particular areas where either in bandwidth or storage that you're going to be building out, or any particular geographical area where you want to increase your CapEx?

Sure. One of the important things about our model is it is a very low CapEx model as compared to most network services businesses because we don't pour concrete, as we say, to build data centers. We don't dig ditches to lay fiber. We don't own dedicated pipes ourselves. Our strategy has been to partner with 1,000 different networks around the world for capacity and efficiency. So the CapEx really falls into a couple buckets. In addition to what JD commented on before, which is to depreciate some of our software developments, we build a service, not a product. That's an important component of investing in R&D. On top of it, most of the network CapEx goes to two places. One is servers. As you know, we buy commodity hardware at the lowest possible price and deploy it. So we continue to add servers to add capacity to the network. The other is in storage devices and that's been a growth area particularly in the M&E space where our media customers want us to store large libraries of content and deliver it. So we're making the incremental investment in low-cost commodity hardware in both cases, either servers or storage devices, and then optimizing their performance through the software and efficiency in the software that we can control and develop uniquely.

Is there any particular geographic area you're going to have to be adding in more of the servers and the storage? Or is it pretty much just across the board?

I think we added two more countries in the past quarter to be over 70 countries now. We just look at where our customers are and where their traffic is going. I would say that we're growing everywhere. Probably Asia is growing disproportionately a little more because of the extent of broadband penetration there. But it's not restricted to any one area, which is another I think very positive sign for the business, that we're seeing strong domestic demand as well as international in Europe and in Asia.

Good enough. Thank you very much, guys.

Our next question comes from Rod Ratliff - Stanford Group.

Hi guys, congratulations. Outstanding job. Every time I think you guys are going to take a quick pause that refreshes with your guidance, you surprise me. So I guess that's a high-class problem to have.

We try to call it straight down the middle of the fairway.

That's a good place to be, in the short grass. Everything pretty much has been asked and answered. I think I would just ask for you to comment a little bit, Paul; there's been a lot of wailing and gnashing of teeth, particularly as we've been out to visit investors with you guys about the difference between what you do and what a BitTorrent type of a solution does for a media and entertainment company. Could you offer up some clarity on how you perceive the Akamai solution to be superior versus a BitTorrent?

I think what you really mean there is what we provide versus an open peer-to-peer solution?

Yes.

You know peer-to-peer technology has been around a lot longer than Akamai has, so there's really nothing new there. There are really a couple of aspects to it. First, our customers are businesses who have a business online. There are a lot of things that are important to them: protecting their content, getting paid for their content, make sure it's delivered to a legitimate user, knowing when it's been delivered, knowing in real-time where that user is to target advertising or specific content. Those are some of the business rules that our network can apply. But in many ways, we run a peer-to-peer network. We have these 20,000 servers in over 70 countries working together. So they are peers of each other. But we just can provide a lot of business rules on top that our customers value and pay us for. Other peer-to-peer solutions don't have those kinds of controls or business value adds and really in an over a decade haven't provided a lot of value for businesses. They're mostly associated with piracy and other bad behavior online and don't really provide the business value. One of the attractions has been this idea of using desktop resources, bandwidth that end-users have paid for as opposed to that a network provides. But again, if you look deep into peer-to-peer, most peer-to-peer solutions wind up going back to an origin server 30% or more of the times that you get no benefit immediately. You know, bad things happen, not just piracy or theft of content. If you're doing a software download and it doesn't complete, you haven't saved anything. You've actually created a problem for your end users. So our view is that we provide business service for businesses. We are agnostic about technology and what kind of technology or network we would use to deliver content as long as we can meet our customers' needs. So far, we haven't seen anything in that peer-to-peer space that meets our customers' needs and is compatible with what they tell us they want. I don't think you've really seen much success from those kinds of technology providers to penetrate the market in a meaningful way.

Great. One other thing. Do you think any of the bondholders might be looking to converting anytime soon, or do you think that they are just going to be fat, dumb and happy?

You know, we haven't seen any conversions since we've been in the super conversion in the premium range, which has been six or eight months now. I think basically, it's kind of trading like equity with a small premium on top of it, plus the interest payments. So I don't anticipate any conversions. But I could be surprised, but I don't anticipate it.

I think the other thing to note is the dilution is already counted. We've already put the shares in. So they can call us anytime they want to, but it won't change our share.

Right. Congratulations again, guys. Thanks a lot.

Our next question comes from Todd Raker - Deutsche Bank.

Hi guys. Nice quarter again. Two questions for you. First, on the competitive landscape, can you give us a feel for who you are seeing competitively? There's been clearly huge uptake here as a market. Who is really trying to penetrate this market? Can you just give us a feel in terms of large competitive bake-offs, have you lost any of the competitive bake-offs in the last three months? If you do lose, why do you lose?

You know, we really don't lose competitively very often, either churn or greenfield opportunities. I think that that's because we've built great technology, a great reputation built on terrific customer service, reliability and scalability. We benefited because I think in the last couple of years, using a content delivery network has become almost a standard part of an RFP for a new web initiative from a big corporation. We would like to think that we're the place they call first, and we're able to satisfy their needs at a price that's very fair and competitive to what our customers require. The basic competitive landscape still remains do it yourself as the first competitor. We rarely meet somebody that says, I don't have any IT business or web business but let me talk to you. They usually are a corporation that already has an IT infrastructure, probably already has some web infrastructure that's trying to decide, is this the time to look for an outsourced solution from a company such as ours. So we have to convince them that we can provide more value at a lower price with lower worries than if they just want to launch another application using their inside resources, which is a function of people, hardware, software, facilities and connectivity or bandwidth. We think we have a very compelling offer. In terms of direct competitors, we still continue to see large-managed service providers with solutions. I always refer to that as the one-rope to choke model. Maybe we don't do any one thing particularly well, but we do lots of things and you can outsource all the worry to us. That's one cluster of competitors, and the other are smaller players who offer point solutions in the content delivery space. Again, we think that we have very compelling offers with a full suite of products and a proven track record. So no fundamental change in what we see out there at all.

Two quick questions incremental to that. I know you guys don't talk specific customers. But if you look at your customer base this past quarter, just trying to get a framework for how diversified it truly is from a revenue perspective. How many customers would you guys estimate did over $1 million in revenue with you guys this past quarter?

We don't give that number, but it was a significant number of customers. We have a group of really stellar customers. But our customer concentration has continued to spread I think every quarter over the last several years. We haven't had a 10% customer in several years. But also if you will, the contribution of the top 5 or the top 10 or the top 20 -- however you would want to cut it -- continues to break in our direction with more and more diversification, both by individual customer name but also largely with verticals tending to spread as well. So we never want to make any customer unhappy or worry about any vertical slowing down for some macro reason. We feel that we're very well distributed and protected.

If you look at that same type of analysis but on the bursting side, if you just looked at the 10 largest events in any one quarter, how material can an event be relative to where you guys are from a revenue run rate to that?

None of them can really be material. They can be large, but it would have to be on an unprecedented scale at this point to be worth more than a couple of percent. If materiality would be 5% or something, we don't come close.

Then, you guys have highlighted the momentum in the media and entertainment vertical. Can you just give a sense for how often are you renegotiating contracts with your customer base there? I assume these guys are kind of blowing through their expectations in terms of requirements.

I would say we tend to negotiate at the end or near the end. But with a large one or somebody who just gets way ahead of their own growth in a positive sense, we are happy to open the contract early. We are happy to extend them a volume discount if they want to step up to more commitment. But we then look for them to extend the length out in the form of a new contract. That happens in the minority of cases, but it's certainly not an exception.

Thanks guys. Great job.

Our next question comes from Erik Zamkoff - Morgan Joseph.

Good afternoon. Most of my questions have been answered. Congrats on a terrific quarter. Just a very quick housekeeping question. Can you give out an ARPU number? I apologize. I've been dancing in between calls.

It was $16,400 which is up 7% quarter-over-quarter.

I was wondering if you could hit on not just during the quarter, you spoke about the BitTorrent competitive scenario. How do you view some of the co-location or data center companies that offer peering? Do you see that as competitive, complementary? Or is it a market that is sustainable on both sides?

It's complementary in the following sense and not competitive. It's complementary in that if our customers can't connect well to the Internet, we can't help them because we have to go back to their site to get content updates, refreshes and if it's a dynamic site to get to the database. So we have always recommended to our customers that they have good connectivity, that it's multi-home, meaning don't have just one pipe to one network. But if you look at the distribution of network traffic across the 15,000 networks that make up the Internet, no single network has more than a low single-digit share of traffic that they deliver. So even if you go into a common carrier or a carrier neutral location and peer with a few networks, you are probably at maximum delivering or sending traffic directly into networks that probably can deal with say one out of five requests. So, it really doesn't solve capacity, security or reliability the way that Akamai does. So it's very compatible, really orthogonal market. That market has tended to be a real estate market. Can you build it and run it for less than you can rent the square footage? Ours is really a value-added services, software-type business with very high margins. Operator, I think we can take maybe two more questions.

Our next question comes from Robert Stimson - WR Hambrecht.

Speedera rolls off this quarter, so I know that a lot of the churn numbers had to do with smaller accounts in terms of average sales price. Going out, can you just give us a sense of churn; is it probably going to be 2.5%, 3% or about the 4% range?

I think you know we said low single digits. We do think that we clearly won't have the Speedera churn that we've had the last couple of quarters. But again, at that low level of churn, the customers that do churn tend to be on the smaller side. It's really insignificant in terms of the impact on our revenue growth.

Then just a follow-up. There was actually an acquisition made by Motorola today in your backyard. I was just curious, it seems like every time we talk to you guys, you come up with kind of a new application doing something with digital radio over the Internet. Motorola is now getting into the video-on-demand business. So I'm just curious, as that technology goes from a host model to video-on-demand and delivering it to the phone, do you play in that market? Would that be an opportunity for Akamai going forward?

I can't comment on a specific customer or prospect, but I think what you're seeing is that media and entertainment is becoming a description of companies you wouldn't have thought about before. So I guess I don't know all the details of today's deal you're talking about. I've just seen the headline. But you might be soon calling a company like this one a media company. Certainly, Apple, you could call a media company in many ways. Five years ago, you wouldn't have thought that way. Those have intended to create incremental opportunities for us to help deliver the content in a more effective way. So you've probably pointed the sales guys in the direction of a new opportunity and we will go look at it.

Great, and then just a final goodbye here. On the number of POPS if you go to your CapEx, can you kind of give us a sense of the number of POPs that are out there in terms of what's in the infrastructure? Then basically as you build out your CapEx spend for the rest of the year this year, how many POPs you guys think you will have worldwide by year end?

The number is over 2,000. We don't give specifics on guidance on where we're going. That's pretty competitive proprietary information.

Congratulations, great job, guys. Sandy, thanks for all of your help.

Thanks. Operator, we will take one more question.

Our next question comes from Brent Bracelin - Pacific Crest Securities.

Thank you. I just really wanted to follow up and talk a little bit more about the growth drivers of the business. Obviously, you've highlighted several kind of areas that are helping drive the acceleration in growth. Obviously, your outlook would suggest you expect that momentum to continue going forward. But, more specifically, could you rank streaming events, software downloads, new customers, an increase in traffic consumption and ARPUs? Of those buckets, what are you seeing that's most surprising? What do you expect to be ranked the most significant drivers going forward? I know you talked about a broad-based recovery. But if you could provide a little more color on what specific areas you expect near-term to be a growth driver, that certainly would be helpful.

That's a tough question because you bundled a whole bunch of things that are interrelated there. I think that the biggest surprise for us has been the dramatic increase that broadband connectivity has played. I think we hit a critical mass a year ago, not just the people who have broadband, but people who really moved to what I would call a broadband lifestyle. Because they have broadband at work and at home, so they can seamlessly go from one experience to another. An online content subscription suddenly had value in more places in their life. So we really got to a tipping point, and it just drove this very strong acceleration in our business. If you just look at the quarter-over-quarter top line increase, Q3 to Q4, we were up 9%, a great result going into our traditionally strongest quarter. Then, we hit Q1 and we went up 10% on a bigger number. And now we went up 11% on an even bigger number moving into the spring, which sometimes has slower traffic growth than the winter months. So it's just been building on itself very nicely and really allowing customers across all of these segments to grow, certainly M&E. That's a very obvious one as sporting events have grown from niche things to audiences that actually are larger than cable networks would like to have in primetime when you have hundreds of thousands of simultaneous users watching an online sporting event. We're seeing it in the pay model of things like music sales online, which has been very strong. We're certainly seeing it in the growth and the distribution of software. As I indicated before, we're seeing it in the growth of our commerce customers, who are just doing more online. There, the traffic growth isn't as large…

This concludes today's conference call. You may now disconnect.